AdWords
3.1K members online now
3.1K members online now
Get started with AdWords - learn the basics to get set up for success
Guide Me
star_border
Reply

Ad rotation and Experiment (beta)

Visitor ✭ ✭ ✭
# 1
Visitor ✭ ✭ ✭

I want to A/B test the text ads of a few ad groups in a campaign. Currently the Ad rotation setting for the campaign is "Optimize for conversions" meaning the impression won't be distributed evenly.

 

If I run Experiment for a selected ad group where the distribution is 50/50, will this override the "Optimize for conversions" setting of the rest of ad groups during the experimental period? 

2 Expert replyverified_user
2 ACCEPTED SOLUTIONS

Accepted Solutions
Marked as Best Answer.
Solution
Accepted by pankaj1782 (Top Contributor)
September 2015

Re: Ad rotation and Experiment (beta)

Top Contributor
# 4
Top Contributor
I agree that configuring experiments  could be confusing: (see my replies in blue)

Now I have 3 questions
1) I wonder if these remaining ad groups will still keep their "Optimize for conversion" setting? - The cost per click method does not change since you do not test it
2) Now I have 2 ad copies A & B under one ad group. I will apply "Control only" for A and "Experiment only" for B. (This ad group is "Control and Experiment") Is this correct? - Correct
3) 50-50 auction split means splitting of impressions that I have. Am I right?- Correct

 

Moshe, AdWords Top Contributor , Twitter | Linkedin | Community Profile | Ad-Globe
Did you find any helpful responses or answers to your query? If yes, please mark it as the ‘Best Answer’

View solution in original post

Marked as Best Answer.
Solution
Accepted by MosheTLV (Top Contributor)
September 2015

Re: Ad rotation and Experiment (beta)

Visitor ✭ ✭ ✭
# 14
Visitor ✭ ✭ ✭
I got the response from Google. The conversion optimization setting of Ad Rotation indeed overrides the 50/50 split rule.
So in conclusion, we should switch the setting to display the text ads evenly to be able to split 50/50 traffic.

View solution in original post

Re: Ad rotation and Experiment (beta)

Top Contributor
# 2
Top Contributor

Hello,

 

As far as I remember, Ad Rotation settings is not actually a part of Experiments. You could try different bids for the keywords, different ad copies and then the Control/Experiment split will decide how often the experiments which you have added are going to apply like 50-50 or any other percentage.

 

This help center reference explains how to actually set up experiments and what all options you have to set them up: https://support.google.com/adwords/answer/2375383

 

Hope this helps!

Pankaj

Re: Ad rotation and Experiment (beta)

Visitor ✭ ✭ ✭
# 3
Visitor ✭ ✭ ✭
Hi Pankaj,

Yes Ad Rotation is not part of Experiments. Currently the campaign setting is still "Optimize for conversion". I want to A/B test 2 different copies for a single ad group, not the whole campaign.

However, after I set up the Experiment to do that, the rest of the ad groups has this "Control and Experiment" sign.

Now I have 3 questions
1) I wonder if these remaining ad groups will still keep their "Optimize for conversion" setting?
2) Now I have 2 ad copies A & B under one ad group. I will apply "Control only" for A and "Experiment only" for B. (This ad group is "Control and Experiment") Is this correct?
3) 50-50 auction split means splitting of impressions that I have. Am I right?
Marked as Best Answer.
Solution
Accepted by pankaj1782 (Top Contributor)
September 2015

Re: Ad rotation and Experiment (beta)

Top Contributor
# 4
Top Contributor
I agree that configuring experiments  could be confusing: (see my replies in blue)

Now I have 3 questions
1) I wonder if these remaining ad groups will still keep their "Optimize for conversion" setting? - The cost per click method does not change since you do not test it
2) Now I have 2 ad copies A & B under one ad group. I will apply "Control only" for A and "Experiment only" for B. (This ad group is "Control and Experiment") Is this correct? - Correct
3) 50-50 auction split means splitting of impressions that I have. Am I right?- Correct

 

Moshe, AdWords Top Contributor , Twitter | Linkedin | Community Profile | Ad-Globe
Did you find any helpful responses or answers to your query? If yes, please mark it as the ‘Best Answer’

Re: Ad rotation and Experiment (beta)

Visitor ✭ ✭ ✭
# 5
Visitor ✭ ✭ ✭

Thanks Pankaj and MosheTLV for clarifying my questions.

 

So it's clear. Ad rotation setting 'Optimize for conversions: Show ads expected to provide more conversions' won't be affected by the experiment.

Re: Ad rotation and Experiment (beta)

Visitor ✭ ✭ ✭
# 6
Visitor ✭ ✭ ✭
I did a check today. Even though I put the 50/50 split rule, the % served of the two ad copies is 80/20.
I really double that the Ad Delivery - Ad Rotation setting of the whole campaign is overriding the experiment rule. Again currently it is "Optimize for conversions: Show ads expected to provide more conversions". Hmm

Re: Ad rotation and Experiment (beta)

Top Contributor
# 7
Top Contributor
Please upload a screenshot of your setting
Moshe, AdWords Top Contributor , Twitter | Linkedin | Community Profile | Ad-Globe
Did you find any helpful responses or answers to your query? If yes, please mark it as the ‘Best Answer’

Re: Ad rotation and Experiment (beta)

Visitor ✭ ✭ ✭
# 8
Visitor ✭ ✭ ✭

I combined three cropped screenshots into one.

 

ad delivery vs experiment beta.png

Re: Ad rotation and Experiment (beta)

Top Contributor
# 9
Top Contributor

Where are the results of the experiments.?

I haven't run an experiment for a while, but you should see the control and "test" separately with an "arrow" showing whether the results are  statistically significant.

Moshe, AdWords Top Contributor , Twitter | Linkedin | Community Profile | Ad-Globe
Did you find any helpful responses or answers to your query? If yes, please mark it as the ‘Best Answer’

Re: Ad rotation and Experiment (beta)

Visitor ✭ ✭ ✭
# 10
Visitor ✭ ✭ ✭

Hi Moshe,

 

This is the screenshot.

 

experiment result.png

 

I wonder if I set up the text ad experiment correctly.

  1. Create a new ad copy
  2. Ad group has both "Control and Experiment" setting
  3. The old ad copy is switched to "Control only". Hence the little sign next to the check box is black.
  4. The new ad group is "Experiment only"

 

So is it correct to say that the campaign Ad rotation/ Delivery setting is overriding the 50/50 rule?

Thanks in advance Smiley Happy